We propose a novel X-ray interferometry imaging system which is suitable for screening mammography. This Multi-Contrast Mammography System with Modulated Phase Gratings (MCMS-MPG) provides 3 complementary types of images ? conventional attenuation, phase-contrast and dark-field (small-angle-scatter) ? all processed from a single scan with an amount of dose delivered to the patient that is comparable to that of a standard screening mammogram. The key innovation that makes this screening interferometry imaging system possible is the modulated phase grating (MPG) which makes high-quality X-ray interference fringes observable by clinical X-ray detectors even at compact distances (<70cm) without the fluence absorbing analyzer-grating required for other advanced X-ray interferometry systems. While the key scientific premise of our system is application independent, in this project we focus on breast cancer application because nearly 1 in 8 women in the United States will develop invasive breast cancer in their lifetime. Screening and early diagnosis are key to reducing the death rate. Nearly 40 million women per year are screened using mammography, a widely accepted radiologic imaging method that uses low-energy X-rays. The current generation of mammography systems yields 92% sensitivity and 84-92% specificity. False positives require recalls for diagnostic imaging and biopsies, the costs of which run to $7.9 billion/year. Improving the sensitivity and specificity for screening will therefore have a significant combined impact on breast-cancer diagnosis as well as related healthcare costs. Our hypothesis is that the additional multi-contrast information that our system provides, obtained with no more dose to patient than with current screening mammograms, will yield higher sensitivity and specificity than those for standard screening mammograms. This project has two aims. Aim1: We will perform comprehensive simulations and develop an iterative recovery method to improve image quality. The recovery algorithm removes unwanted image degradation effects like incoherent scatter. The simulator and recovery algorithm will be validated experimentally using digital breast phantoms and an observer study by radiologists. Aim2: We will build a prototype version of our MCMS-MPG system and evaluate its performance on human breast pathology samples and mammography quality assurance phantoms, using radiation doses comparable to screening mammography systems. An observer study with radiologists will be used for this evaluation.